python-pillow/Pillow
1
1
Fork 0
mirror of https://github.com/python-pillow/Pillow.git synced 2025-02-25 08:10:33 +03:00
Code Issues Packages Projects Releases Wiki Activity

issue #2959: change Py3 detection, fix trailer location for some PDFs

Browse Source
This commit is contained in:
Dvořák Václav 2018-01-24 22:45:37 +01:00
parent cfacf8bff8
commit 991f832d91
1 changed files with 19 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

41
src/PIL/pdfParser.py
View File

@ -11,11 +11,11 @@ except ImportError:
UserDict = collections.UserDict UserDict = collections.UserDict
if sys.version_info.major >= 3: if str == bytes: # Python 2.x
make_bytes = lambda s: s # pragma: no cover
else: # Python 3.x
def make_bytes(s): def make_bytes(s):
return s.encode("us-ascii") return s.encode("us-ascii")
else:
make_bytes = lambda s: s # pragma: no cover
def encode_text(s): def encode_text(s):
@ -141,20 +141,20 @@ class PdfName():
allowed_chars = set(range(33,127)) - set((ord(c) for c in "#%/()<>[]{}")) allowed_chars = set(range(33,127)) - set((ord(c) for c in "#%/()<>[]{}"))
def __bytes__(self): def __bytes__(self):
if sys.version_info.major >= 3: if str == bytes: # Python 2.x
result = bytearray(b"/")
for b in self.name:
if b in self.allowed_chars:
result.append(b)
else:
result.extend(make_bytes("#%02X" % b))
else:
result = bytearray(b"/") result = bytearray(b"/")
for b in self.name: for b in self.name:
if ord(b) in self.allowed_chars: if ord(b) in self.allowed_chars:
result.append(b) result.append(b)
else: else:
result.extend(b"#%02X" % ord(b)) result.extend(b"#%02X" % ord(b))
else: # Python 3.x
result = bytearray(b"/")
for b in self.name:
if b in self.allowed_chars:
result.append(b)
else:
result.extend(make_bytes("#%02X" % b))
return bytes(result) return bytes(result)
__str__ = __bytes__ __str__ = __bytes__
@ -212,17 +212,14 @@ class PdfBinary:
def __init__(self, data): def __init__(self, data):
self.data = data self.data = data
if sys.version_info.major >= 3: if str == bytes: # Python 2.x
def __bytes__(self):
return make_bytes("<%s>" % "".join("%02X" % b for b in self.data))
def __str__(self):
return bytes(self).decode("us-ascii")
else:
def __str__(self): def __str__(self):
return "<%s>" % "".join("%02X" % ord(b) for b in self.data) return "<%s>" % "".join("%02X" % ord(b) for b in self.data)
else: # Python 3.x
def __bytes__(self):
return make_bytes("<%s>" % "".join("%02X" % b for b in self.data))
def pdf_repr(x): def pdf_repr(x):
if x is True: if x is True:
@ -239,7 +236,7 @@ def pdf_repr(x):
return bytes(PdfDict(x)) return bytes(PdfDict(x))
elif isinstance(x, list): elif isinstance(x, list):
return bytes(PdfArray(x)) return bytes(PdfArray(x))
elif isinstance(x, str) and sys.version_info.major >= 3: elif isinstance(x, str) and str != bytes:
return pdf_repr(x.encode("utf-8")) return pdf_repr(x.encode("utf-8"))
elif isinstance(x, bytes): elif isinstance(x, bytes):
return b"(" + x.replace(b"\\", b"\\\\").replace(b"(", b"\\(").replace(b")", b"\\)") + b")" # XXX escape more chars? handle binary garbage return b"(" + x.replace(b"\\", b"\\\\").replace(b"(", b"\\(").replace(b")", b"\\)") + b")" # XXX escape more chars? handle binary garbage
@ -396,9 +393,9 @@ class PdfParser:
whitespace_mandatory = whitespace + b"+" whitespace_mandatory = whitespace + b"+"
newline_only = br"[\r\n]+" newline_only = br"[\r\n]+"
newline = whitespace_optional + newline_only + whitespace_optional newline = whitespace_optional + newline_only + whitespace_optional
re_trailer_end = re.compile(whitespace_mandatory + br"trailer" + whitespace_mandatory + br"\<\<(.*\>\>)" + newline \ re_trailer_end = re.compile(whitespace_mandatory + br"trailer" + whitespace_optional + br"\<\<(.*\>\>)" + newline \
+ br"startxref" + newline + br"([0-9]+)" + newline + br"%%EOF" + whitespace_optional + br"$", re.DOTALL) + br"startxref" + newline + br"([0-9]+)" + newline + br"%%EOF" + whitespace_optional + br"$", re.DOTALL)
re_trailer_prev = re.compile(whitespace_optional + br"trailer" + whitespace_mandatory + br"\<\<(.*?\>\>)" + newline \ re_trailer_prev = re.compile(whitespace_optional + br"trailer" + whitespace_optional + br"\<\<(.*?\>\>)" + newline \
+ br"startxref" + newline + br"([0-9]+)" + newline + br"%%EOF" + whitespace_optional, re.DOTALL) + br"startxref" + newline + br"([0-9]+)" + newline + br"%%EOF" + whitespace_optional, re.DOTALL)
def read_trailer(self): def read_trailer(self):
search_start_offset = len(self.buf) - 16384 search_start_offset = len(self.buf) - 16384